TrajDeleter: Enabling Trajectory Forgetting in Offline Reinforcement Learning Agents

arXiv:2404.12530v1 Announce Type: new
Abstract: Reinforcement learning (RL) trains an agent from experiences interacting with the environment. In scenarios where online interactions are impractical, offline RL, which trains the agent using pre-collected datasets, has become popular. While this new paradigm presents remarkable effectiveness across various real-world domains, like healthcare and energy management, there is a growing demand to enable agents to rapidly and completely eliminate the influence of specific trajectories from both the training dataset and the trained agents. To …

abstract agent agents arxiv become cs.cr cs.lg datasets domains enabling energy environment healthcare interactions new paradigm offline paradigm popular reinforcement reinforcement learning the environment trains trajectory type world